Ollama
A local runtime for downloading, running, and serving open language models through a simple CLI and API.
Overview
Ollama packages model weights, a runtime, and a local API so that open language models can be run on a workstation or server. It handles model downloads and keeps inference local unless the operator exposes the service.
Where it fits
- Running local chat and coding models
- Prototyping applications against a local model API
- Keeping prompts and outputs on owned hardware
Key capabilities
- Model pull and run commands
- Local HTTP API
- GPU acceleration where supported
- Modelfiles for custom prompts and parameters
Repository metadata
- Stars (snapshot)
- 185,988
- Main language
- Go
- License
- MIT
- Topics
- deepseek, gemma, glm, go, golang, gpt-oss, llama, llm, llms, minimax, mistral, muse, ollama, qwen
- Category
- Developer tools
- Last push
- 2026-11-05T01:32:18Z
Repository description from GitHub metadata: Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Official links
- Official repository
- Read the README The README is linked, not copied or cached by this site.
Leaving this site. These links open github.com in a new tab.
Source archives, release assets, checksums, and license texts are hosted and controlled by
GitHub and the project maintainers. Review the upstream license before redistributing or
modifying anything.